zhangjy328
zhangjy328
@uncle-tou @hhu-aiboy 你好，我跑别的模型也出现了这个问题，请问你们是怎么解决的？我也是ubuntu系统跑的
I also really want to know what the structure of the training dataset is like. Has anyone successfully run the training? Could you share the dataset directory?
I want to train an image dataset of rectangular shapes, but I don't want to perform any size changes. Is that okay?
And, Do I have to train only one type of sample each time? I tried to include multiple types of samples in the training set simultaneously, but the loss value...
我也遇到了类似的问题 ValueError: Unrecognized keyword arguments passed to Conv2D: {'weights': [array([[[[-0.19253726, -0.08509053, 0.00574696, ..., -0.13854688,
> 在sceneflow上，我论文中默认是训练20w次迭代。如果你不需要那么多迭代的话，我建议可以10w次迭代，可以以10w次迭代做消融实验。 作者你好， 我用的是kitti数据集，总共就400个数据，训练20w次迭代，不会过拟合吗？另外为什么kitti数据集只用10结尾的数据，不用11结尾的数据？
Hello, I have also encountered the same problem. Can we understand that the maximum depth is a conversion ratio scale from relative distance to absolute distance? If I know the...
> 我成功从主分支安装了 PyTorch3D 0.7.8，无需修改任何代码。 > > **环境：** > > * Windows 11 > * CUDA 12.8 > * PyTorch 2.7.1+cu128 > * Python 3.9.23 （Anaconda） > * Visual Studio 2022...
@nhansendev Hello, I encountered the same problem when I tested the pix2pix model. When I trained it normally, it was ok, but when I ran down the code: ` python...